Cylindroporella sugdeni Elliott, 1957, an Early Cretaceous Middle Eastern Dasycladalean alga – a revision

The type material of Cylindroporella sugdeni Elliott, 1957, is revised. The total lack of sterile laterals and consequently the absence of the diagnostic feature of the genus Cylindroporella, i.e., the alternation of sterile and fertile laterals within the same whorl, leads to its exclusion from the genus Cylindroporella Johnson, 1954, and to its ascription to the genus Holosporella Pia, 1930. Recent stratigraphic investigations suggest that the stratum typicum originally reported as Early Cretaceous in age is more precisely Hauterivian or possibly Early Barremian in age.


Introduction
Except for the fossil species ascribed to the fossil family Seletonellaceae in which the lateral expansions are randomly distributed along the alga main axis, the Dasycladalean algae have laterals arranged in regular rows that are called "verticils".The latter comprise either sterile laterals or fertile laterals, except for the representatives of Cylindroporella Johnson, 1954, and Korkyrella Sokač & Velić ex Sokač, 2004, both ascribed to the fossil family Triploporellaceae.Having their sterile and fertile laterals occurring in alternation within the same row, both genera have a unique body plan amidst the fossil and living Dasycladalean algae.The more their representatives are reassessed, the more one questions the existence of such a body plan (see Barattolo and Parente, 2000, for instance).
Following the description of its type species, Cylindroporella barnesii Johnson, 1954, two new species were introduced by Elliott (1957): the Early Cretaceous C. sugdeni and the Late Jurassic C. arabica.The revision of the second species listed, i.e., C. sugdeni Elliott, 1957, is addressed below.

Material and methods
Thanks to a loan organized by the Department of Palaeobotany of the Natural History Museum, London, the five thin sections that correspond to the type material of Cylindroporella sugdeni Elliott, 1957, which are currently stored in the Collection PAL (Palaeontology) under numbers PB V 41620 to 41624 (Fig. 1), were examined.The thin section labels indicate that the material comes from a depth of 4380', i.e., 1335 m (measured depth, MD), from core 14 in well Fahud 1 in Oman.
In addition to this material, another thin section, which is part of a set given by Graham F. Elliott to Jacques Emberger (Université de Bordeaux), was also studied.The label of this thin section implies it comes from the well "Fahud 1 (Oman), core 14, depth 4381' (MD)".This last thin section is currently deposited with a LPB (Laboratoire de Paléontologie de Brest) number in the collections of the Département des Sciences de la Terre et de l'Univers, Université de Bretagne Occidentale, Brest (France).
The dominant texture is a bioclastic grainstone with micritized grains of all sizes (peloids to aggregates, micritic extraclasts) and a few ooids.The skeletal grains consist mostly of the new alga, indeterminate lituolid and textulariid Published by Copernicus Publications on behalf of The Micropalaeontological Society.foraminifers, rare Praechrysalidina infracretacea Luperto Sinni, 1979, and echinoderm remains.Unfortunately, the thin sections are rather thick (probably thicker than 30 µm).The original piece of limestone was porous but it was not injected with resin in order to consolidate it.As a consequence the material is locally crumbled (with local overlaps) or was partly removed.In addition, small air bubbles are trapped in the Canada balsam.In order to better understand the alga structure, a photo editor was used to color in blue the holes in the thin section views (Fig. 2); then some visible features (Fig. 3), possibly belonging to the alga, were contoured.

Stratigraphical background
The name of the Lower Cretaceous group changes across the international boundary.In the United Arab Emirates (UAE,  (Elliott, 1957, pl. 1, fig. 6); (c-d) transverse section, paratype (one of the syntypes), thin section PB V 41623 (Elliott, 1957, pl. 1, fig. 3).The parts colored in blue correspond to parts removed while making the thin section and possibly some original porosity.White arrows pointing to the upper right indicate air bubbles and those pointing to the lower left indicate overlaps in the thin section manufacturing; opposing red arrows indicate an induced fracture.All photos on the same scale (scale bar = 500 µm).
among which is the emirate of Abu Dhabi) and in Saudi Arabia, the equivalent Lower Cretaceous unit is called the Thamama Group (Fig. 4).Core 14 with Cylindroporella sugdeni Elliott, 1957, was retrieved from an interval labeled as "Habshan".For decades, oil and gas company geologists have developed and used an interregional stratigraphic framework for this interval (e.g., Hassan et al., 1975).According to Granier (2000;Granier et al., 2003), most of their operational units prove to be unconformity-bounded units.In addition, because they can also be characterized by their discrete fossil contents, Granier (2008) treats them as genuine regional stages.In Abu Dhabi, the shallow-water carbonate units of the Thamama Group comprise, from top to bottom, the Bab and the Shu'aiba (Upper Aptian), the Hawar (Lower Aptian), the Kharaib (Upper Barremian-Lower Aptian), the Lekhwair (Upper Valanginian-Hauterivian-Lower Barremian), the Zakum (Lower Valanginian), the Belbazem (Berriasian), the Bu Haseer (transition of the Tithonian to the Berriasian), and the Habshan (Tithonian).However, in Oman, the shallow-water carbonates of the lower part of the interval may be missing or they may pass to hemipelagic and pelagic facies of the Kahmah Group (Granier, 2000(Granier, , 2008;;Granier et al., 2003).In addition to the UAE units, two facies-driven formations were originally described from the Kahmah Group in Oman: the Rayda and the Salil (e.g., see Hassan et al., 1975).The Rayda, which is also an unconformity-bounded unit, is time equivalent to both the Bu Haseer and Belbazem (Tithonian-Berriasian).The Salil facies is the basinal time equivalent to the rest of the stratigraphic column (Valanginian-Aptian pars).Accordingly, there are not only geopolitical reasons but also geological reasons (ages, hiatuses, and development of basinal facies) that may justify the use of the Thamama Group in the UAE and that of the Kahmah Group in Oman.

Stratigraphic location of the type level
In well Fahud 1, the lithostratigraphic succession identified by Petroleum Development Oman (PDO) and Shell geologists comprises, from top to bottom, the "Shu'aiba" (that includes the Hawar), the Kharaib, the "Lekhwair", the "Habshan", and the Salil (Fig. 4).The use of quotation marks is justified by the fact that the definition of these units does not match the regional standardized scheme (Hassan et al., 1975;Granier 2000Granier , 2008;;Granier et al., 2003).This is especially true in the case of the Habshan of the well Fahud 1, which refers to a transitional zone between the Salil basinal facies below and Lekhwair facies above and which has no connection with the type Habshan in Abu Dhabi (latest Jurassic in age).In conclusion, because the type level of Cylindroporella sugdeni Elliott, 1957, is located in the median part of an interval equivalent to the type Lekhwair (it is not located in the Kharaib as erroneously stated in Granier, 2008), and although an Early Barremian age cannot be excluded, it is most probably Hauterivian in age (Fig. 4).
1954, which is based on Cylindroporella barnesii Johnson, 1954.Today, due to some recent revisions, the number of its representatives has significantly diminished (Table 1).
According to Sokač (2004), sterile laterals of Korkyrella texana Johnson ex Sokač, 2004, are rather long and thin primaries; the fertile laterals consist of ovoid primaries followed by short secondaries.According to Conrad (1982), sterile laterals of Cylindroporella barnesii Johnson, 1954 consist of rather long and thin primaries followed by short secondaries; the fertile laterals are ovoid primaries.However, Cylindroporella sugdeni Elliott, 1957 does not match any of these diagnostic criteria.There are not any long and thin laterals, supposedly sterile, but only ovoid primaries without secondaries, supposedly fertile.In tangential sections, the empty spaces remaining between adjacent ovoid primaries could have been confused with sterile primaries.In this case, the algal sections may recall those of Montiella elitzae (Bakalova, 1971), a form with laterals consisting of a thin primary bearing a thin secondary and an ovoid am-pulla at its outer end.Nonetheless, several photomicrographs clearly attest the lack of fertiles.Consequently, the species is transferred to the genus Holosporella (Pia, 1930), which is characterized by a simple body plan with a main axis bearing closely set verticils of vesicular laterals, commonly arranged in a quincunx from one verticil to the next.(Elliott, 1957, pl. 1, fig. 2), and 3.

Phylum
Original diagnosis: (Elliott, 1957) "Large, thick cylindroporelliform segments, agreeing in internal structure with those of Cylindroporella barnesii, but with greater diameter of central canal and sporangia and externally, and a greater distance between whorls; it is proportionally shorter and much thicker than the type species (. . .)".
Description: Thalli are found in the form of broken calcareous tubes with an axial cavity in lieu of the main axis and globular pores in lieu of the laterals.The inner diameter of such a calcareous coating commonly represents one-fourth of its outer diameter.The calcification may vary from one specimen to the others or within a single specimen.Usually it forms a thin crust around the laterals and the main axis.Micritic infills may be observed above and beyond this crust (Fig. 5b right, g and k right) but in some cases the "intercellular" space, i.e., the empty space remaining  between the vesicular laterals and the main axis, is filled in by a drusy calcite cement (Fig. 5a and l).Previously, such intercellular micritic infills with ill-defined contours were erroneously interpreted as thin sterile laterals.
No unquestionably cyst-like structures were observed but the laterals are supposedly fertile because their volume is significantly larger than that of the axis cavity, which excludes that the species is endosporous (with cysts located in the main axis) and supports it being interpreted as cladosporate (with cysts located in the primary laterals).The proximal part of these laterals consists of a short and narrow peduncle (Fig. 5c, h-j and k left) that expands abruptly into the characteristic wider vesicular part.
Ovoid laterals from successive verticils have a quincunx arrangement (Fig. 6a) that results from the imbrication of the verticils, clearly visible in tangential sections (Fig. 5a, b right, g, k right and l).The imbrication ratio is computed by the difference between the maximum width of a lateral, i.e., the thickness of a verticil, and the distance between two successive verticils at their insertion points, divided by this last distance.Imbrication is particularly strong in this species in which it represents nearly two-thirds of the whorl spacing.
Emended diagnosis: Rather large Holosporella species with a relatively narrow main axis and strongly imbricated verticils of wide vesicular laterals, supposedly fertile.Using the various measurements listed below (D, d, d/D, h, etc.) may help discriminate it from the other representatives of the genus.In (a) the five X's at the center of the pores illustrate the quincunx arrangement of the laterals.Key for the measurements is as follows.L: maximum length observed; D: outer diameter; d: inner diameter; h: whorl spacing; e: thickness of the calcareous coating; α: angle of the pore to the main axis; l: length of the primary pores; p: visible width of a primary pore; pp: width of the proximal part of the primary pores; pd: maximum width of the primary pores.
Lectotype: The lectotype (Figs.3l and 5c) selected here is the oblique section of pl. 1, fig. 2 of Elliott (1957) from the thin section labeled PB V 41621.
Remark: Some features observed in axial sections are not visible in transverse sections and vice versa.Oblique sections should always be preferred as a type (holotype, lectotype, neotype) to axial and transverse sections because a palaeophycologist usually gathers more information from oblique sections than from any axial or transverse section.

Conclusions
Holosporella sugdeni Elliott, 1957, nov. comb., is known with certainty at least from the Lekhwair and Kharaib of Abu Dhabi (Granier et al., 2003) and from the Jezzinian of Lebanon (Basson and Edgell, 1971), i.e., from the Hauterivian and Barremian interval in the Middle East.More than half a century after it was introduced by Elliott (1957), there are few correct identifications of this species.Records outside this region commonly come with "cf." or "aff." and do not expressly refer to this species but most probably to Montiella elitzae (Bakalova, 1971).Thus, Holosporella sugdeni Elliott, 1957, nov. comb., might be an endemic species.Following the revision of this species, the genus Cylindroporella Johnson, 1954, has lost another representative.As Barattolo and Parente (2000) did earlier, one can question the existence of its unique body plan with sterile and fertile laterals occurring in alternation within the same row.Further investigations on this fossil algal genus are requested, requiring the revision of its remaining representatives and its type species.Oman (PDO) and Shell, and particularly Michael Darke, Gordon Forbes and Iain Prince for having shared some proprietary information on well Fahud 1 with me; I wish to acknowledge the help of Michael Quentin Morton, who acted as a facilitator.I am also indebted to the Natural History Museum, London, and specifically to Lee S. Davies and Lil Stevens for having organized the loan of Elliott's original material.I am grateful to Filippo Barattolo and Geraint Wyn Hughes for having reviewed the paper and suggesting helpful improvements.Last but not least, special thanks are due to Phil Savador for having polished the English text of a previous version of this paper.
Edited by: Francesca Sangiorgi Reviewed by: Filippo Barattolo and Geraint Wyn Hughes

Figure 1 .
Figure 1.The set of six thin sections studied.Both sides of 41623 and 41624 are visible; the back sides of them were rotated in order to make the text readable.

Figure 2 .
Figure 2. (a-b) Oblique section, paratype (one of the syntypes), thin section PB V 41624(Elliott, 1957, pl. 1, fig.6); (c-d) transverse section, paratype (one of the syntypes), thin section PB V 41623(Elliott, 1957, pl. 1, fig. 3).The parts colored in blue correspond to parts removed while making the thin section and possibly some original porosity.White arrows pointing to the upper right indicate air bubbles and those pointing to the lower left indicate overlaps in the thin section manufacturing; opposing red arrows indicate an induced fracture.All photos on the same scale (scale bar = 500 µm).

Figure 4 .
Figure 4. Stratigraphic framework of the United Arab Emirates and Oman (modified fromGranier et al., 2003;Granier, 2008;Granier and Busnardo, 2013; * here the Late Aptian includes the Furcata ammonite Zone).The large star indicates the supposed relative location of the core sampling.

Figure 6 .
Figure6.Reconstructions (a) of a heavily calcified Holosporella and (b) of a weakly calcified one.In (a) the five X's at the center of the pores illustrate the quincunx arrangement of the laterals.Key for the measurements is as follows.L: maximum length observed; D: outer diameter; d: inner diameter; h: whorl spacing; e: thickness of the calcareous coating; α: angle of the pore to the main axis; l: length of the primary pores; p: visible width of a primary pore; pp: width of the proximal part of the primary pores; pd: maximum width of the primary pores.

Table 1 .
Johnson, 1954.changes that have affected more than half of the former representatives of the genus CylindroporellaJohnson, 1954.